See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/10773229
`
`Lipid evaluation in HIV-1-positive patients treated with protease inhibitors
`
`READS
`82
`
`Andre Dejam
`Harvard Medical School
`
`55 PUBLICATIONS   7,491 CITATIONS   
`
`SEE PROFILE
`
`Article  in  Antiviral Therapy · February 1999
`
`Source: PubMed
`
`CITATIONS
`45
`
`8 authors, including:
`
`Georg Behrens
`Hannover Medical School
`
`222 PUBLICATIONS   11,841 CITATIONS   
`
`SEE PROFILE
`
`Michael P Manns
`Hannover Medical School
`
`2,966 PUBLICATIONS   108,289 CITATIONS   
`
`SEE PROFILE
`
`Some of the authors of this publication are also working on these related projects:
`
`Clinical View project
`
`Human adult liver stem cells generated from patient samples View project
`
`All content following this page was uploaded by Andre Dejam on 23 May 2014.
`
`The user has requested enhancement of the downloaded file.
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1012, p. 1 of 9
`
`

`

`Antiviral Therapy 4: 163–170
`
`Lipid evaluation in HIV-1-positive patients treated
`with protease inhibitors
`Hartmut H-J Schmidt1, Georg Behrens2*, Janine Genschel1, Matthias Stoll2, André Dejam2,
`Regina Haas1, Michael P Manns1 and Reinhold E Schmidt2
`
`1Department of Gastroenterology and Hepatology and 2Department of Clinical Immunology, Medizinische Hochschule Hannover,
`D-30623 Hannover, Germany
`
`*Corresponding author: Tel: +49 511 532 6656; Fax: +49 511 532 5648; E-mail: Behrens.Georg@MH-Hannover.de
`
`that human
`is accumulating evidence
`There
`immunodeficiency virus type 1 (HIV-1) protease
`inhibitors (PIs) can induce hyperlipidaemia. To evaluate
`the frequency and type of hyperlipidaemia in PI-treated
`patients, 98 outpatients were prospectively analysed for
`their lipoprotein characteristics at the Medizinische
`Hochschule in Hannover, Germany. Fifty-seven percent
`of the patients studied presented with hyperlipidaemia.
`Both hypertrigylceridaemia (type IV and V hyperlipopro-
`teinaemia, 33%) and hypercholesterolaemia (type IIa
`hyperlipoproteinaemia, 6%) were detectable. The
`remaining 18% had a type IIb hyperlipoproteinaemia.
`Increased lipid levels were highly statistically signifi-
`cant compared to a control group of PI-naive
`HIV-1-infected patients [low-density lipoprotein (LDL)
`146 mg/dl (range, 53–274 mg/dl) versus 105 mg/dl
`(range, 22–188 mg/dl; P=0.0006); very-low-density
`lipoprotein (VLDL) 35.5 mg/dl (5–253 mg/dl) versus 18
`
`mg/dl (range, 3–94 mg/dl; P=0.0002)]. All PIs used
`(saquinavir, indinavir, nelfinavir and ritonavir) were
`associated with this variable form of hyperlipidaemia
`according to the Fredrickson classification. There was
`no significant correlation of any determined lipid value
`with the duration of treatment. A higher frequency of
`the apolipoprotein E2 allele and E4 allele was observed
`in the hyperlipidaemic subjects. Patients with excessive
`hypertriglyceridaemia showed a reduced lipoprotein
`lipase activity. Lipodystrophy was observed especially in
`hyperlipidaemic patients and to a lesser extent in
`normolipidaemic subjects. The frequency of hyperlipi-
`daemic risk factors was surprisingly high in the group
`studied, which in turn may explain the proposed
`increased risk of atherogenesis in HIV-1 PI-treated
`patients. Therefore, PI-treated subjects should also be
`evaluated for their lipoprotein pattern, which may
`require antihyperlipidaemic interventions.
`
`Introduction
`
`The early introduction of powerful protease inhibitors
`(PIs) in the treatment of human immunodeficiency
`virus type 1 (HIV-1)-infected patients has resulted in
`unexpected side effects over time [1–10]. The HIV-1
`protease consists of two 99 amino acid subunits that
`form a homodimer with a typical retroviral aspartyl
`protease activity [11]. PIs bind with high affinity to the
`catalytic site of HIV-1 protease and prevent cleavage of
`HIV precursor proteins in infected cells. In the presence
`of PIs, the virion is unable to mature and the amount
`of infectious virus rapidly decreases. The four common
`PIs used (indinavir, nelfinavir, ritonavir and saquinavir)
`are structurally related substances that inhibit the
`cleavage of HIV precursor proteins competitively [11].
`There is accumulating evidence that PIs can induce
`hyperlipidaemia in a subgroup of treated patients
`[5,12]. Patients reported to date have presented with
`hypertriglyceridaemia. In addition, there are also rare
`
`events of coronary artery and peripheral vascular
`disease published so far [13–15]. The accelerated
`atherogenesis described in these subjects may result
`from the observed dyslipidaemia. To further evaluate
`the lipid pattern and its potency as an atherogenic risk
`factor we analysed PI-treated subjects in detail.
`
`Patients and Methods
`
`Patients
`HIV-1-infected patients who were admitted to our HIV
`outpatient clinic from March to July 1998 were
`enrolled in the study and analysed for lipid abnormal-
`ities. Specimen collection followed a 12 h overnight
`fasting period. Lipid parameters of patients receiving
`PI therapy were compared with a control group of PI-
`naive patients [13 patients treated with two nucleoside
`reverse transcriptase inhibitors (NRTI) and/or non-
`
`©1999 International Medical Press 1359-6535/99/$17.00
`
`163
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1012, p. 2 of 9
`
`

`

`HH-J Schmidt et al.
`
`Table 1. The clinical characteristics of the various hyperlipoproteinaemic phenotypes using the Fredrickson classification are
`depicted
`
`Phenotype
`Type IV/V
`Type IIa
`Type IIb
`Normal
`
`n, number of patients.
`BMI, body mass index.
`
`n
`32
`6
`18
`42
`
`Age (years)
`44.2±10.6
`54.2±6.6
`53.4±11.4
`45.9±12.7
`
`BMI
`24.3±4.1
`24.6±2.4
`24.7±3.0
`24.0±3.5
`
`CD4 cells (cells/mm3)
`294.8±186.7
`281.5±169.0
`287.2±159.2
`314.2±189.8
`
`Viral load (copies/ml)
`52.1×103±88.3×103
`2.3×103±3.7×103
`56×103±14×103
`35.8×103±12.3×103
`
`Lipodystrophy (%)
`28
`33
`28
`7
`
`nucleoside reverse transcriptase inhibitor (NNRTI) and
`10 therapy-naive patients]. None of the patients
`received parenteral nutrition, nutrition via percuta-
`neous gastrostomy, growth hormones or anabolic
`steroids. Body fat changes in patients with HIV infec-
`tion were defined as signs of abnormal
`fat
`accumulation (dorsocervical fat pad, benign symmetric
`lipomatosis, increased abdominal girth/visceral obesity,
`breast hypertrophy) and/or loss of the subcutaneous fat
`in the face, limbs, gluteal regions or buttocks. The
`symptoms were defined clinically by physical examina-
`tion and patient’s reports. The different forms of
`hyperlipoproteinaemia were defined as total choles-
`terol >200 mg/dl, triglycerides >200 mg/dl, low-density
`lipoprotein (LDL) >155 mg/dl, and very-low-density
`lipoprotein (VLDL) >35 mg/dl. Reference values for
`other lipid parameters were high-density lipoprotein
`(HDL) >35 mg/dl and lipoprotein(a) [Lp(a)] >30 mg/dl.
`
`Determination of triglycerides and total cholesterol
`Serum concentrations of triglycerides and total
`cholesterol were determined using semi-automatic
`enzymatic methods. For the triglyceride assay the
`Peridochrom Triglyceride GPO-PAP Test kit was used
`(Boehringer Mannheim, Germany). The reaction
`buffer
`contained 0.5 mM ATP, 0.35 mM
`4-aminophenazone, 3 U/ml lipase, 2.5 U/ml glycerol
`phosphatoxidase, 0.2 U/ml glycerol kinase, 0.15 U/ml
`peroxidase and 3.5 mM 4-chlorophenol. For the total
`cholesterol assay the CHOD-PAP Test kit (Boehringer
`Mannheim, Germany) was used. The reaction buffer
`contained 50 mM MgCl2, 1 mM 4-aminophenazone,
`10 mM sodium cholate, 6 mM phenol, 4 mM 3,4-
`dichlorophenol, 0.3% polyglycol ether, 0.4 U/ml
`
`cholesterol esterase, 0.25 U/ml cholesterol oxidase, 0.2
`U/ml peroxidase. Extinction was measured at 546 nm.
`The assays were performed according to manufac-
`turer’s instructions.
`
`Lipid electrophoresis of VLDL, LDL and HDL
`The concentrations of VLDL-, LDL- and HDL-choles-
`terol were determined using an electrophoretic and
`densitometric method (REP Lipoprotein-Kit, Helena
`Diagnostics, Germany). The plasma lipoproteins were
`separated on an agarose gel according to the manufac-
`turer’s instructions. The gel was stained and scanned in
`a specific densitometer. The lipoprotein concentrations
`were automatically calculated according to their
`densitometric value. Hyperlipoproteinaemia was
`classified according to the Fredrickson classification.
`
`Nephelometric quantification of apolipoproteins
`[apoA-I, apoA-II, apoB, apoE and Lp(a)]
`This method is based on particle agglutination tech-
`nology. The assays were semi-automatically performed
`on the BN II (Behring Nephelometer, Behringwerke
`AG, Marburg, Germany) using rabbit polyclonal anti-
`sera against human apolipoproteins. For Lp(a), the
`antibody is conjugated to latex-particles to enhance the
`turbidity of the antigen–antibody complexes [16]. The
`turbidity of the formed complexes was proportional to
`the concentration of the used antigen. The reactions
`were measured at time 0 and 6 minutes at 840 nm. The
`differences of the measured values were calculated
`automatically. This approach excludes non-specific
`turbidity. From each of the serum samples, 30 µl was
`diluted by 1:100. A lyophilized serum control and stan-
`dard serum were used for the quantification. The
`
`Table 2. The lipoprotein parameters of the various hyperlipoproteinaemic phenotypes are illustrated
`
`Phenotype
`Type IV/V
`Type IIa
`Type IIb
`Normal
`
`Serum cholesterol
`249.7±74.9
`268.7±30.5
`298.2±42.5
`184.9±39.3
`
`Serum triglycerides
`654.3±542.4
`169±40.8
`425.6±201.9
`145±67.6
`
`VLDL
`84.6±56.4
`25±7.8
`65.6±34.1
`20.2±8.6
`
`LDL
`139.4±36.6
`203.5±22.1
`188.9±37.4
`122.6±30.9
`
`HDL
`27±9.2
`40.2±9.7
`33.4±8.2
`42.2±14.6
`
`Lipoprotein (a)
`(mg/dl)
`19.9±30.9
`26.8±27.3
`21.2±42.9
`10.8±14.5
`
`VLDL, very-low-density lipoprotein.
`LDL, low-density lipoprotein.
`HDL, high-density lipoprotein.
`
`164
`
`©1999 International Medical Press
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1012, p. 3 of 9
`
`

`

`Lipid evaluation in HIV-1-positive individuals
`
`Table 3. Lipid parameters, CD4 cell count and viral load of PI-treated and PI-naive HIV-infected patients
`
`Cholesterol (mg/dl)
`Triglycerides (mg/dl)
`LDL (mg/dl)
`VLDL (mg/dl)
`HDL (mg/dl)
`ApoA-I (mg/dl)
`ApoA-II (mg/dl)
`ApoB (mg/dl)
`ApoE (mg/dl)
`CD4 cell count (cells/mm3)
`HIV RNA copies/ml (log10)
`
`PI-treated patients
`227 (100–450)
`252.5 (58–2192)
`146 (53–274)
`35.5 (5–253)
`31.5 (15–91)
`135 (93–189)
`30 (18–43)
`125 (2–33)
`5 (2–33)
`275 (5–814)
`3.20 (1.69–5.88)
`
`PI-naive patients
`159.5 (75–307)
`125.5 (22–538)
`105 (22–188)
`18 (3–94)
`36 (20–50)
`140 (95–238)
`29 (22–54)
`84 (53–180)
`4 (2–10)
`336 (123–652)
`3.097 (1.69–4.72)
`
`P value
`<0.00001
`0.0003
`0.0006
`0.0002
`0.3
`0.8
`0.2
`0.0004
`0.001
`0.02
`0.4
`
`Data are expressed as median (range) and P values are indicated for between-group comparison. VLDL, very-low-density lipoprotein cholesterol; LDL, low-density
`lipoprotein cholesterol; HDL, high-density lipoprotein cholesterol
`
`reaction buffer consisted of polyethylene glycol,
`sodium chloride (11.6 g/l) in 0.05 M phosphate buffer
`pH 7.0 and sodium azide (<1 g/l). N-Diluens is a phos-
`phate-buffered saline, pH 7.5, containing sodium azide
`(<1 g/l). Samples were measured in duplicate.
`
`Lipoprotein lipase and hepatic lipase activity
`Lipoprotein lipase (LPL) activity and hepatic lipase
`(HL) activity were measured according to the methods
`described in [17].
`
`ApoE genotype
`DNA was extracted from patients blood using a
`DNA purification kit
`(Invitek, Germany).
`Purification was performed according to the manu-
`facturer’s instructions. The relevant region within the
`apoE gene was amplified by PCR. PCR products
`were digested with the restriction enzyme HhaI for 4
`h at 37°C. After digestion, samples were analysed on
`a 3% Nusieve, 1% Seakem agarose gel (FMC
`Bioproducts, USA) was used to detect the different
`apoE alleles.
`
`CD4 cell count and viral load
`CD4 lymphocyte cell count was determined by flow
`cytometry and HIV RNA copies by quantitative PCR
`(Amplicor HIV-1 Monitor Test Kit, Roche Molecular
`Systems).
`
`Statistical analysis
`The analysis included the calculation of the mean and
`standard deviation (SD). Statistical significance for
`lipid profiles between PI-treated patients and control
`group was determined by analysis with non-para-
`metric Mann–Whitney U test or Student’s t-test where
`appropriate. Differences between the PI regimens and
`types of hyperlipoproteinaemia were determined by
`analysis of variance (ANOVA). P values of <0.05
`were considered to indicate statistical significance.
`
`Results
`
`Ninety-eight HIV-1-positive patients treated with PIs
`(78 male, 20 female) were analysed for lipid abnor-
`malities. None of the studied subjects had overt
`diabetes mellitus. The analysis of the lipid pattern
`revealed that 56 (57%) had detectable hyperlipidaemia
`showing that the prevalence of hyperlipidaemia in
`HIV-1 PI-treated subjects was very high. In addition,
`10 patients (19%) presented with an elevated Lp(a)
`serum concentration (range, 33–158 mg/dl, mean
`81±41 mg/dl), which represents an additional cardio-
`vascular risk factor. Interestingly, seven out of these 10
`patients with increased Lp(a) levels were treated with
`indinavir. In contrast, increased levels for Lp(a) were
`observed in 7% of the 42 normolipidaemic subjects
`(range, 44–71 mg/dl, mean 54±15 mg/dl) and in two
`patients (9%) of the control group (range, 44–100
`mg/dl, mean 77 mg/dl).
`We also determined the type of hyperlipopro-
`teinaemia using the Fredrickson classification [18]. A
`hypertriglyceridaemic lipid pattern was detectable in
`32 subjects, three of them presented with type V hyper-
`lipidaemia (3%) and 29 of them with type IV
`hyperlipidaemia (30%). Eighteen subjects had an
`increase of both VLDL- and LDL-cholesterol, the so-
`called type IIb hyperlipidaemia (18%). In addition, six
`subjects had an isolated hypercholesterolaemia, type
`IIa hyperlipidaemia (6%). Thus, besides hypertriglyc-
`eridaemia, PI-treated HIV-1-positive patients can also
`develop isolated hypercholesterolaemia. The clinical
`characteristics of the various hyperlipoproteinaemic
`phenotypes are depicted in Table 1, whereas Table 2
`reflects their lipoprotein parameters. Since type V
`hyperlipidaemia is commonly an exacerbation of an
`existing type IV hyperlipidaemia, we grouped these
`affected subjects for the subsequent statistical analysis.
`The average body mass index (BMI) was calculated in
`all studied groups within the non-obese range (<25
`
`Antiviral Therapy 4:3
`
`165
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1012, p. 4 of 9
`
`

`

`type.
`To determine the impact of the apoE gene on lipid
`patterns in the studied subjects, we evaluated the geno-
`type, illustrated in Figure 1. Since apoE 3/3 is the
`normal and most common genotype in individuals, the
`analysis revealed a relatively higher association of the
`apoE 2 and statistically significant higher rate of apoE
`4 allele with patients presenting with hyperlipopro-
`teinaemia. Moreover, carriers of the apoE 4 allele had
`higher total triglyceride values than patients without
`the apoE 4 allele (mean 534.0±606.3 mg/dl versus
`321.9±278.0 mg/dl, P=0.02). No significant differ-
`ences were observed for the other lipid parameters. The
`percentages of each apoE genotype compared to the
`frequency of apoE genotype of three different German
`population studies [19–21] revealed higher frequencies
`of apoE 2/3 (19.8% versus 10.7–12.0%), apoE 2/4
`(7.3% versus 1.5-3.0%), and apoE 4/4 (8.3% versus
`1.3–2.8%) in our study group.
`The duration of treatment with PIs was compa-
`rable
`between
`the
`normolipidaemic
`and
`hyperlipidaemic groups: median 19 months (range,
`2–30) in the normolipidaemic group, whereas median
`21.5 months (range, 17–29 months), median 18.5
`months (range, 5–29 months), median 20 months
`(range, 2–28 months), and median 16 months (range,
`2–26 months) in type IIa, type IIb and type IV/V
`hyperlipidaemic subjects, respectively. Since the onset
`of hyperlipidaemia might already be within several
`months after initiation of PI treatment, the duration
`of therapy seems not to be critical for triggering
`hyperlipidaemia. The patients in the major groups of
`PI therapy were on drugs for an equivalent time
`[saquinavir: median 14 month (range, 7–30), indi-
`navir: median 17 month (range, 1–29), nelfinavir:
`21.5 month (range, 3–30), ritonavir plus saquinavir:
`21 month (range, 2–26)]. There was no significant
`correlation of any determined lipid value with the
`duration of PI treatment. Other individual factors
`have to be responsible for the development of hyper-
`lipidaemia. The individually applied PIs were also not
`significantly correlated with both hypertriglyceri-
`daemia
`(VLDL-cholesterol)
`and
`hypercholesterolaemia (LDL-cholesterol) (Figures
`2–4). More importantly, the combination of two PIs
`(especially ritonavir plus saquinavir) seems to have an
`synergistic effect on lipid metabolism as the preva-
`lence of hyperlipidaemia is particulary high in these
`groups (Figure 2). These data reflect that all PIs used
`are associated with hyperlipidaemia, which in turn
`can occur in variable phenotypes. Figure 5 illustrates
`as an example, the lipoprotein profile in patients
`under the treatment with indinavir, reflecting the vari-
`able lipoprotein phenotype once hyperlipidaemia is
`detectable.
`
`HH-J Schmidt et al.
`
`Figure 1. The apoE allele (ApoE2, E3 and E4) and
`corresponding hyperlipoproteinaemia
`
`Hyperlipidaemic patients
`
`Normolipidaemic patients
`
`25
`
`20
`
`15
`
`10
`
`Number of patients
`
`5
`
`0
`
`E2
`
`E3
`ApoE alleIe
`
`E4
`
`96 of 98 patients were evaluated.
`
`kg/m2). The CD4 cell count and the viral load in all
`groups were within the same range. Interestingly,
`lipodystrophy was physically detectable in both hyper-
`lipidaemic and normolipidaemic subjects. However,
`lipodystrophy was more commonly observed among
`the hyperlipidaemic patients.
`The statistical evaluation of patients receiving PI
`compared to the PI-naive group (20 male, three female)
`revealed highly statistically significant differences
`(Table 3). PI treatment was associated with elevated
`levels for fasting cholesterol (median, 227 mg/dl;
`range, 100–450 mg/dl; P<0.00001), triglycerides
`(median, 252.5 mg/dl; range, 58–2192 mg/dl;
`P=0.0003), LDL (median, 146 mg/dl; range, 53–274
`mg/dl; P=0.0006), VLDL (median, 35.5 mg/dl; range,
`5–253 mg/dl; P=0.0002). However, HDL values were
`within the same range in PI-treated (median, 31.5
`mg/dl; range, 15–91 mg/dl) and PI-naive patients
`(median, 36 mg/dl; range, 20–50 mg/dl; P=0.3).
`The additional quantification of apolipoprotein A-I
`(apoA-I), apolipoprotein B (apoB), apolipoprotein A-II
`(apoA-II), and apolipoprotein E (apoE) did reveal
`significant higher levels of apoB (P=0.0004) and apoE
`(P=0.001) in the PI-treated group (Table 3), which is
`characteristic for patients with increased levels of
`VLDL- and/or LDL-cholesterol. The determination of
`the LPL and HL activity in eight patients with
`increased serum trigylcerides above 800 mg/dl revealed
`that in seven out of eight patients the mean LPL
`decreased by 30% (range 20–70%) in comparison to
`control subjects, whereas the hepatic lipase activity
`was normal in the eight subjects studied. LPL from PI-
`treated patients was compared to normolipidaemic
`HIV-negative control subjects with an apoE 3/3 geno-
`
`166
`
`©1999 International Medical Press
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1012, p. 5 of 9
`
`

`

`Figure 2. The presence of hyperlipidaemia or
`normolipidaemia of patients treated with the different
`HIV-1 PIs
`
`Figure 3. The VLDL-cholesterol serum concentrations of
`patients treated with different HIV-1 PIs
`
`Lipid evaluation in HIV-1-positive individuals
`
`300
`
`250
`
`200
`
`150
`
`100
`
`50
`
`0
`
`Very-low-density lipoprotein cholesterol (mg/dl)
`
`Hyperlipidaemic
`patients
`
`Normolipidaemic
`patients
`
`Ritonavir + nelfinavir
`
`Ritonavir + indinavir
`
`Ritonavir + saquinavir
`
`Saquinavir + nelfinavir
`
`Ritonavir
`
`Nelfinavir
`
`Indinavir
`
`Saquinavir
`
`No protease inhibitor
`
`20
`
`10
`
`0
`Number of patients
`
`10
`
`20
`
`Saquinavir
`
`Indinavir
`
`Nelfinavir
`
`Ritonavir
`
`Saquinavir + nelfinavir
`
`Ritonavir + saquinavir
`
`Ritonavir + indinavir
`
`Ritonavir + nelfinavir
`
`Discussion
`
`There is accumulating evidence that PIs induce hyperlip-
`idaemia in a subgroup of patients, which has been
`recently reported [5,12,22]. SoRelle summarized these
`initial findings, which were discussed at the 12th World
`AIDS Conference in Geneva, highlighting the benefit of
`this drug regimen and its potential adverse effects [12].
`Both atherosclerosis and pancreatitis have been observed
`in association with hyperlipidaemia in PI-treated patients
`[23]. We analysed 98 HIV-1-infected subjects for their
`lipid parameters confirming the increased frequency of
`hyperlipidaemia in PI-treated patients. The observed
`hyperlipidaemia is primarily an increase of the serum
`concentration of the triglyceride-rich VLDL fraction
`causing a type IV hyperlipoproteinaemia [18]. However,
`patients may also develop isolated hypercholesterolaemia
`(type IIa hyperlipoproteinaemia), combined hypercholes-
`terolaemia and hypertriglyceridaemia
`(type
`IIb
`hyperlipoproteinaemia), and excessive hypertriglyceri-
`daemia (type V hypertriglyceridaemia), respectively.
`Metabolic alterations, e.g. hypertriglyceridaemia, have
`been described already in HIV-infected individuals espe-
`cially in the late phase of the disease and to a lesser extent
`as we found in the PI-treated group. The major causes of
`hypertriglyceridaemia were shown to be elevated rates of
`de novo lipogenesis and delayed clearance of triglycerides
`in the postprandial period [24]. Interestingly, in our study
`both PI-treated as well as PI-naive patients had low
`plasma contents of HDL, already known to occur early
`in HIV disease [24]. Thus HDL seems not to be affected
`by PI treatment. The onset of the variable lipid patterns
`
`described and associated with PI-containing drug regi-
`mens has
`similarity with
`familial
`combined
`hyperlipidaemia, a lipid disorder, which is inherited with
`variable penetrance and variable type of hyperlipopro-
`teinaemia [25]. The primary origin of both PI-induced
`hyperlipidaemia and familial combined hyperlipidaemia
`has not been elucidated so far. The presence of variable
`phenotypes of hyperlipidaemia in PI-treated patients may
`reflect the impact of multiple genetic and metabolic
`factors.
`The use of PIs may result in hypertriglyceridaemia,
`but also in lipodystrophy [5,7,8]. The latter represents
`partial lipodystrophy, which is known to be frequently
`associated with hypertriglyceridaemia [26]. Metabolic
`in vivo analysis of VLDL particles revealed that the
`origin of hypertriglyceridaemia in patients with partial
`lipodystrophy is due to their delayed catabolism [27]
`and increased production [28,29]. Thus these studies
`will have also to be performed on PI-treated patients in
`the future to determine the aetiology of hyperlipid-
`aemia. The cause of delayed catabolism of the triglyc-
`eride-rich VLDL particles can be caused by a reduced
`activity of the LPL activity and HL activity, respec-
`tively. In addition, postprandial hyperinsulinaemia,
`which is associated with PI treatment [22], is able to
`induce increased hepatic lipogenesis leading to the
`release of VLDL particles. We determined the lipase
`activities in selected studied patients presenting with
`excessive hypertriglyceridaemia. These data revealed
`that seven out of eight subjects had reduced LPL
`
`Antiviral Therapy 4:3
`
`167
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1012, p. 6 of 9
`
`

`

`HH-J Schmidt et al.
`
`Figure 4. The LDL-cholesterol levels of patients treated with
`different HIV-1 PIs
`
`Figure 5. This figure reflects the occurence of different types
`of hyperlipidaemia under the treatment of indinavir
`
`25
`
`20
`
`15
`
`10
`
`5
`
`0
`
`Number of patients
`
`300
`
`250
`
`200
`
`150
`
`100
`
`50
`
`0
`
`Low-density lipoprotein cholesterol (mg/dl)
`
`Normal
`
`HLP
`type IIa
`
`HLP
`type IIb
`
`HLP
`type IV
`
`HLP
`type V
`
`may reflect the polygenicity of this mechanism.
`Although the low-density lipoprotein receptor-
`related protein (LRP) and the cytoplasmatic retinoic
`acid binding protein type 1 (CRABP-1) have been
`suggested as potential targets for PIs, explaining the
`variable clinical features including hyperlipidaemia of
`treated patients, no proven data are available so far
`[33]. Proteases are in general very important in
`lipoprotein metabolism including lipoprotein assembly,
`secretion, catabolism and cell membrane receptor
`interactions. ApoB and apoA-I are the main structural
`proteins for LDL and HDL particles e.g. are proteolyt-
`ically modified intracellularly before the secretion of
`the lipoprotein particles [34–36]. After secretion, the
`proapoA-I is proteolytically cleaved to result into the
`mature apoA-I. We assume that within the blood,
`mature apoA-I is also degraded by proteases, which is
`essential for determining the HDL serum concentration
`[37]. ApoA-I has evolutionary similarities with e.g.
`apoE and apoA-II [38–40]. Therefore, proteolysis will
`presumably play also an essential role in the structural
`and enzymatic functions of apoE and apoA-II. In addi-
`tion, LCAT and CETP modulate as enzymes the
`lipoprotein metabolism, which in turn are activated by
`the cofactors apoA-I, apoA-IV, ApoC-I, and apoD,
`respectively [41,42]. Receptor-mediated cellular uptake
`of lipoproteins may also serve as a potential target,
`especially since PIs may also act intracellularly. Thus
`PIs have many potential targets in modulating the
`lipoprotein metabolism.
`The determination of the lipoprotein profile in PI-
`treated patients is mandatory. Depending on the
`severity of the hyperlipidaemia and the benefit of the PI
`used, its continuation or withdrawal has to be decided.
`
`Ritonavir + nelfinavir
`
`Ritonavir + indinavir
`
`Ritonavir + saquinavir
`
`Saquinavir + nelfinavir
`
`Ritonavir
`
`Nelfinavir
`
`Indinavir
`
`Saquinavir
`
`No protease inhibitor
`
`activity, but normal HP activity. LPL activity is affected
`by different functional defective mutations [17]. It is
`possible that patients with genetic variants leading to
`less catabolic activity of the enzyme are more suscep-
`tible for the development of PI-associated hypertri-
`glyceridaemia. The discontinuation of PIs seems to
`result in the normalization of lipoprotein parameters
`[30,31], reflecting a potential impact of these drugs on
`the LPL activity. In addition, the metabolism of VLDL
`and LDL is influenced by the apoE genotype.
`Interestingly, we found a relatively higher association
`of the apoE 2 and apoE 4 allele in the patients with
`hyperlipidaemia [32]. Therefore, the apoE genotype
`may play a role in the pathogenesis of hyperlipidaemia
`in these patients.
`We also determined the serum concentration of
`Lp(a), an atherogenic risk factor, which is primarily
`regulated by its hepatic synthetic rate. Nineteen per
`cent of our hyperlipidaemic patients had an additional
`pathological increased serum concentration of Lp(a)
`emphasizing the increased risk of atherogenic events.
`These patients were primarily treated with indinavir
`(7/10), yet the real implication of this observation
`remains unclear. More patients have to be studied to
`assess the relevance of these data.
`The long-term treatment with PIs does not seem to
`affect the severity of hyperlipidaemia. Thus this drug
`regimen may induce hyperlipidaemia depending on the
`genetic background such as the mentioned apoE geno-
`type. All PIs used are associated with hyperlipidaemia
`and the type of hyperlipidaemia can differ. This in turn
`
`168
`
`©1999 International Medical Press
`
`Hikma Pharmaceuticals
`
`IPR2022-00215
`
`Ex. 1012, p. 7 of 9
`
`

`

`To circumvent drug withdrawal, we have to gain addi-
`tional experience in potential treatment strategies for
`hyperlipidaemia to minimize the risk of atherogenesis
`and pancreatitis.
`
`Acknowledgements
`
`We thank Ulrike Beisiegel for the determination of the
`LPL activity and HL activity in a subgroup of the
`studied patients. Hartmut H-J Schmidt and Georg
`Behrens both contributed equally to the study design,
`interpretion of data and drafting of the article.
`
`References
`
`2.
`
`4.
`
`7.
`
`1. Dube MP, Johnson DL, Currier JS & Leedom JM. Protease
`inhibitor-associated hyperglycaemia. Lancet 1997;
`350:713–714.
`Eastone JA & Decker CF. New-onset diabetes mellitus
`associated with use of protease inhibitor. Annals of
`Internal Medicine 1997; 127:948.
`3. Hengel RL, Watts NB & Lennox JL. Benign symmetric
`lipomatosis associated with protease inhibitors. Lancet
`1997; 350:1596.
`Lumpkin MM. Reports of diabetes and hyperglycaemia in
`patients receiving protease inhibitors for the treatment of
`human immunodeficiency virus. FDA, Public Health
`Advisor June 11, 1997.
`5. Carr A, Samaras K, Burton S, Law M, Freund J, Chisholm
`DJ & Cooper DA. A syndrome of peripheral lipodys-
`trophy, hyperlipidaemia and insulin resistance in patients
`receiving HIV protease inhibitors. AIDS 1998;
`12:F51–F58.
`6. Dong BJ, Gruta C, Legg J, Balano K & Goldschmidt G.
`Diabetes and the use of protease inhibitors. 5th Conference
`on Retroviruses and Opportunistic Infections, Chicago,
`USA, 1–5 February 1998. Abstract 416.
`Lo JC, Mulligan K, Tai VW, Algren H & Schambelan M.
`“Buffalo hump” in men with HIV-1 infection. Lancet
`1998; 351:867–870.
`8. Miller KD, Jones E, Yanovski JA, Shankar R, Feuerstein I
`& Falloon J. Visceral abdominal-fat accumulation associ-
`ated with use of indinavir. Lancet 1998; 351:871–875.
`9. Mulligan K, Tai VW, Algren H, Chernoff DN, Lo JC &
`Schambelan M. Evidence of unique metabolic effects of
`protease inhibitors. 5th Conference on Retroviruses and
`Opportunistic Infections, Chicago, USA, 1–5 February
`1998. Abstract 414.
`10. Viraben R & Aquilina C. Indinavir-associated lipodys-
`trophy. AIDS 1998; 12:F37–F39.
`11. Flexner C. HIV-protease inhibitors. New England Journal
`of Medicine 1998; 338:1281–1292.
`12. SoRelle R. Vascular and lipid syndromes in selected HIV-
`infected patients. Circulation 1998; 98:829–830.
`13. Behrens G, Schmidt H, Meyer D, Stoll M & Schmidt RE.
`Vascular complications associated with use of HIV
`protease inhibitors. Lancet 1998; 351:1958.
`14. Meyer D, Behrens G, Schmidt RE & Stoll M.
`Osteonecrosis of the femoral head in patients receiving
`HIV protease inhibitors. AIDS 1999; 13:1147–1148.
`15. Henry K, Melroe H, Huebsch J, Hermundson J, Levine C,
`Swensen L & Daley J. Severe premature coronary artery
`disease with protease inhibitors. Lancet 1998; 351:1328.
`16. Schmidt HH, Genschel JC, Wagner S & Manns MP.
`Quantification of lipoprotein(a): comparison of an auto-
`mated latex-enhanced nephelometric assay with an
`
`Lipid evaluation in HIV-1-positive individuals
`
`immunoenzymometric method. European Journal of
`Clinical Chemistry and Clinical Biochemistry 1996;
`34:119–124.
`17. Brümmer D, Evans D, Berg D, Greten H, Beisiegel U &
`Mann WA. Expression of type III hyperlipoproteinemia in
`patients homozygous for apolipoprotein E-2 is modulated
`by lipoprotein lipase and postprandial hyperinsulinemia.
`Journal of Molecular Medicine 1998; 76:355–364.
`18. Fredrickson DS & Lees RS. A system for phenotyping
`hyperlipoproteinemia. Circulation 1965; 31:321–327.
`19. Utermann G, Steinmetz A & Weber W. Genetic control of
`human apolipoprotein E polymorphism: comparison of
`one- and two-dimensional techniques of isoprotein
`analysis. Human Genetics 1982; 60:344–351.
`20. Menzel HJ, Kladetzky RG & Assmann G. Apolipoprotein
`E polymorphism and coronary artery disease.
`Arteriosclerosis 1983; 3:310–315.
`21. Luley C & Wieland H. Epidemiologie der
`Dyslipoproteinämien. In Handbuch der
`Fettstoffwechselstörungen. Edited by P Schwandt and WO
`Richter. 1995; pp 84–103. Stuttgart, New York: Schattauer.
`22. Behrens G, Dejam A, Schmidt H, Balks H-J, Brabant G,
`Körner T, Stoll M & Schmidt RE. Impaired glucose toler-
`ance, beta cell function and lipid metabolism in
`HIV-patients under treatment with protease inhibitors.
`AIDS 1999; 13:F63–F70.
`23. McBride M, Chin MTM, Lee MS & Kelly B. Lipid
`lowering therapy in patients with HIV infection. Lancet
`1998; 352:1782–1783.
`24. Grunfeld C, Pang M, Doerrler W, Shigenaga JK, Jensen P
`& Feingold KR. Lipids, lipoproteins, triglyceride clearance,
`and cytokines in human immunodeficiency virus infection
`and the acquired immunodeficiency syndrome. Journal of
`Endocrinology and Metabolism 1992; 74:1045–1052.
`25. de Graaf J & Stalenhoef AF. Defects of lipoprotein metabo-
`lism in familial comined hyperlipidemia. Current Opinon
`in Lipidology 1998; 9:189–196.
`26. Peters JM, Barnes R, Bennett L, Gitomer WM, Bowcock
`AM & Garg A. Localization of the gene for familial partial
`lipodystrophy (Dunnigan variety) to chromosome 1q21-22.
`Nature Genetics 1998; 18:292–295.
`27. Enzi G, Digito M, Baldo-Enzi G, Cominacini L, Dodi G,
`Carraro R & Zurlo F. Lipid metabolism in lipatrophic
`diabetes. Hormone and Metabolism Research 1988;
`20:587–591.
`28. Chait A, Janus E, Mason AS & Lewis B. Lipodystrophy
`with hyperlipidemaemia: the role of insulin in very low
`density lipoprotein over-synthesis. Clinical Endocrinology
`1979;10:173-178.
`29. Franklin B, Ginsberg H, Haque WU, Yeh HC, Horlick
`MN, Paterniti JR, Gibson J, Le AN & Ginsberg-Fellner F.
`Very